Junseok Park

This study examines cerebellar volume in extremely low birth weight infants and finds that age-adjusted volumetric patterns correlate with developmental outcomes.

This paper introduces a vision-based model for predicting CNV pathogenicity using track-based spatial representations, achieving improved performance on benchmarks and clinical datasets.

scDeepVariant improves germline variant calling in scRNA-seq by integrating population allele frequency data, outperforming existing methods in precision, recall, and rare variant detection.

RetroNet detects low-mosaicism somatic mobile element insertions using deep learning on sequencing read images, achieving high precision and recall with minimal reads.

The paper develops a framework to detect and trace somatic retrotransposition events in human tissues, identifying rare L1 insertions and improving detection accuracy.

This study shows that ATM deficiency causes microglial activation, which contributes to neurodegeneration in Ataxia-Telangiectasia by inducing cytotoxicity in neurons.

The paper identifies rare somatic mutations in sporadic ALS and FTD cases, suggesting they contribute to disease and widespread neurodegeneration.

The paper introduces an interactive retrieval system for clinical trial protocols that improves search by using context-dependent protocol element selection.

This study compares mucosal microbiota composition and function along the intestine using shotgun metagenomics, highlighting functional enrichments and antimicrobial resistance patterns.

The paper develops a participant-driven data collection platform for evaluating digital healthcare apps, demonstrating its reliability through statistical validation with 718 participants.